*Jari-Mikko Meriläinen
*Jyväskylä University School of Business and Economics

*Replication data and Stata codes for the study:
*The role of gender in hate speech targeting politicians: evidence from Finnish Twitter

use hatespeech.dta

tab ministerName, sum(hateful)

*Regressions
logit hateful i.female logfollowers age i.partyid i.leader i.primeminister i.dayofweek
logit hateful i.female##c.logfollowers age i.partyid i.leader i.primeminister i.dayofweek
logit hateful i.female age i.partyid leader primeminister i.dayofweek

*The role of minister portfolio allocation
logit hateful i.female i.portfolio age logfollowers i.partyid i.dayofweek
logit hateful i.female##i.portfolio age logfollowers i.partyid i.dayofweek

*Marginal effect
logit hateful i.female##c.logfollowers age i.partyid i.dayofweek
margins, dydx(i.female) at(logfollowers=(8.019284(0.2)12.38839)) atmeans

*Stereotyping tweets by gender
tab female stereotype if stereotype == 1
tab female stereotype_gender if stereotype_gender == 1
tab female stereotype_political if stereotype_political == 1
tab female stereotype_immigrant if stereotype_immigrant == 1
tab female stereotype_other if stereotype_other == 1